DC Ammeter

You remind me of a sign I saw at a small independent machine shop. I don't remember the sign verbatim but it went something like this.

Here at XYZ machining we proudly provide our customers:

1. High quality work
2. Fast turn around time
3. A low cost

The customer can choose no more than two of the above items.

To have a chance to help you in your search, you must provide more details. About how much current are you reading and at what voltage? How many digits of accuracy do you need and desire? (These are rarely the same.) If this is not a DC current, what frequency is the AC current? I do not make a press machine and to my knowledge Radio Frequency is not used in a press machine. Please define the abbreviation "RF".

the current range is from 0 to 5 amps DC. i would like accuracy to the thousandths but hundredths would work to. the machine is High Frequency heat sealing machine. 20Kw. it is used in making medical blankets.

Well you could do it with a moving coil analogue DC meter, but you would need to do the math and monitor the needle movement. A cheap and fairly accurate solution for basic periodic monitoring.

The other option is a better handheld multimeter which should have the additional functionality you need (digital storage, averaging, peak hold, etc), but you may need a rather expensive one to get what you want.

You did not provide a voltage. However this will probably work.

_{I do not work for, nor have I used any Laurel electronics.}

Wait a minute, 20kW with only 5A of current. That implies a 4kV voltage. Something is not right here.

It might just be a bias current in the LV section and that's a guess as well.

Tell us more about your existing DC ammeter.

Is it a voltmeter connected across a shunt or is it a DC clamp meter pickup?

You are trying to read amps with a digital meter which wont stabilise due to the hf interference from your machine.digital meters do not like hf radio transmissions

You will have to use an analog meter as previously suggested, with a large scale you could read tenths of an amp, hundredths might be a bit tight.

One solution might be to use an oscilloscope and a calibrated shunt resistor (say .1Ω). Newer scopes allow you to find averages, peaks, and a lot of other parameters of a waveform in digital readout.

i am using a anolog meter 0-5 amps that shows tenths. what i am measuring is the plate current that runs through my die. i do not see the full 20kw of the RF. i have the RF sheiled so that the digital gauge will not be affected (so i am told by the meter manufacture). the ammeter is across a shunt. this meter will be placed in a panel perminatly so hand held meters is not an option. i was thinking of an ocilliscope but did not want such a big piece of equipment to have to manage.

The shunt and the analog meter movement combined makes an ammeter. You must remember this when you try to replace this with a digital meter. Now there are digital ammeter assemblies that also use a shunt. The company I referred to earlier also make this. Unlike the "doom and gloom" nay sayers here, one can integrate a digital meter in the same box with a puny 20kw RF source. Unfortunately for you though I cannot provide anything but broad statements for you on how one does this. You will have to experiment and do several measurements along the way but it can be done.

I recommend getting a digital version of the shunt ammeter. In parallel with this shunt you will have to install a low inductance low leakage capacitor. The R of the shunt and the C of this capacitor will create a high pass filter network that provides a RF return path though your DC supply and not your digital read head. I would prefer that this high pass filter would have a -3db corner no higher that two decades below the RF frequency. If possible locate a digital reading head that is isolated from Earth. The input impedance of the two leads of this display will likely be very high. So each lead can handle a low pass network to Earth.

Now to give you the values of all of these capacitors, resistors and possibly inductors I would have to get from you considerably more information on your system. I may even have to visit your facility to do my own testing. I would also have to give you a bill for my services. But I can guarantee that what you want can be done.

Sometimes there's a place for a good analog meter still.

Eh?

Stu.

I do not believe for a single second, that the shielding is adequate for instrumentation at all. It may be adequate for people to work around it.

You need to open up the shielding container to put the plastic into it to be welded. Any contacts, that are frequently moved, are subiect to contact deterioration, hence shield deterioration, that you may not notice at all. The solution for that is a handheld radiation meter made for microwave ovens. Cheap, easy to replace, repeatable reading. It does not matter, that it was designed for an entirely different frequency band. At 20 kW source power it will register. Take reading at correctly set up shields, then you know the deterioration later on. Stray fields may bother people, but drive instrumentation crazy.

Now, back to the DC (or AC?) ammeter. Forget digital. They will go crazy in RF. A large face, relatively insensitive old electromechanical one is the best starting point. For starter look at www.mouser.com/ modutec and www.mouser.com/simpson It lists AC current transformer and Shunts if needed. Pick up the shunt's signal from whenever via a shielded multiwire cable. Ground the shield only at the instrumentation housing's shield. Ground the housing somewhere on the same machine. Have the meter's face shielded with some fine wire mesh too. The metering is insulated from the housing. Input filter: a few tens of milliHenry in series of both legs, 10 nanoFarads across. That ought to remove any RF difference between the legs, permitting correct measurement of the DC component. Mark the GO / NoGo on the face of the meter. If you have a need for data collection, do it via RF module fit for the system employed at the factory. Ask the maker or provider. They will be glad to service you on that one.

It will take a single resistor to set the scale right on the meter. A few trials will get it right for you. And that is the precision you will get in real life.

Ok, what about an analog swinging arm meter/recorder? I still see them on equipment that has intermittent inductive loads like motors with startup currents etc. It would give you a running graph on a paper disk and show you spikes and dips quite well, yet it will not have the jitters that a digital recorder would. You get the advantages of a storage meter and the advantages of an old-school moving coil.

Sorry I don't have a good circuit drawing package handy. I'll try to put it in words. Get a high impedance digital voltmeter. Determine its input impedance (resistance) either from the specs or by measuring. You can measure it by reading a battery and then putting various resistors in series until the reading is 1/2 as much. Get a resistor 9X the meter resistance to put in series. It will now read 1/10 the voltage. Put a capacitor across the meter. The capacitor size determines the time constant you want, along with the series resistor (R x C = Time). The voltmeter reading should be (1/10) x Shunt Resistance x Current

There are digital multimeters, integrating at interval controlled from the outside. As I remember, one of Solartron multimeters have such an option. It can also collect samples and calculate average. So You can have integrals (current(t)) of single pulses of current, collected in memory, up to 500 pulses, at least. With possibility to calculate avg, min, max (math built in multimeter). You can also control mesurements/transmit results to computer via IEEE 488. Probably accuracy of the order of 0.01% should be possible (excluding influence of RF - should be filtered off).

National Instruments cards should allow also to do this. Or dedicated meter, designed to this purpose.

You need to measure it to make research (temporary), or meter should be installed as part of equipment for end user?